A pathway toward enamel-inspired damage-tolerant ceramics

Abstract

Mineralized tissues, including nacre, bone, and teeth, have long been a source of inspiration for novel engineering ceramic materials due to their exceptional fracture resistance. Their performance is derived from complicated, hierarchical microstructures that enable a variety of toughening mechanisms not available in conventional ceramics. Improvements in fracture toughness up to 5x greater than that of the base mineral are not uncommon. Dental enamel is one such system. Teeth are composed of ~96% mineral yet can remain in service with cracks present despite a highly corrosive and wear-intense service environment. However, attempts to implement enamel-inspired structures in synthetic materials have been stymied by the lack of a quantitative description of enamel microstructure. To establish a viable approach towards enamel-inspired materials, this dissertation seeks to answer the following questions: (1) Can printing defects be used to mimic the weak interfaces observed in biological materials? If so, (2) what are the specific, quantitative features in the enamel microstructure that must be mimicked? And (3), can those observed structures be translated into a pattern printable with current AM methods? To answer those questions, the microstructure of enamel from several mammalian species was investigated using microCT imaging to establish a novel, quantitative description of the enamel microstructure. From this description, a printable model using filament extrusion additive manufacturing technologies was established and validated against the quantitative microstructural data. This model capitalizes on residual defects often present in additive manufactured parts which were shown to guide fracture and improve fracture toughness in a vat-polymerization printed zirconia ceramic. This approach is designed to allow the printed parts to mimic the mechanical behavior of enamel, not just the structural arrangement, by introducing the array of weak interfaces characteristic of enamel into the synthetic mimic. Finally, the development of a machine-learning algorithm to enable the generation of a larger microstructural dataset from SEM of enamel is described. This algorithm will help address the future work indicated by this dissertation.